Anda di halaman 1dari 10

Original Research Article

Accepted: April 22, 2011

Published online: June 29, 2011

Dement Geriatr Cogn Disord 2011;31:379387

DOI: 10.1159/000328970

Early Post-Stroke Cognition in Stroke

Rehabilitation Patients Predicts
Functional Outcome at 13 Months
JrgenWagle b,d,e LasseFarner b,d,e KjellFlekky e TorgeirBruunWyller b,e
LeivSandvik c BrynjarFure b BrynhildStensrd b KnutEngedal a,e

Centre for Ageing and Health, Norwegian Centre for Dementia Research, b Department of Geriatric Medicine and
Centre for Clinical Research, Oslo University Hospital, Ullevaal, d Specialist Service in Psychiatry, Norwegian
Centre for Ageing and Health, Vestfold Hospital Trust, Tnsberg, and e Faculty of Medicine, University of Oslo,
Oslo, Norway

Key Words
Cognition Functional outcome Stroke

Objective: To identify prognostic factors associated with
functional outcome at 13 months in a sample of stroke rehabilitation patients. Specifically, we hypothesized that cognitive functioning early after stroke would predict long-term
functional outcome independently of other factors. Methods: 163 stroke rehabilitation patients underwent a structured neuropsychological examination 23 weeks after hospital admittance, and their functional status was subsequently evaluated 13 months later with the modified Rankin
Scale (mRS) as outcome measure. Three predictive models
were built using linear regression analyses: a biological model (sociodemographics, apolipoprotein E genotype, prestroke vascular factors, lesion characteristics and neurological stroke-related impairment); a functional model (pre- and
early post-stroke cognitive functioning, personal and instrumental activities of daily living, ADL, and depressive symptoms), and a combined model (including significant variables, with p value !0.05, from the biological and functional
models). Results: A combined model of 4 variables best pre-

2011 S. Karger AG, Basel

Fax +41 61 306 12 34

Accessible online at:

dicted long-term functional outcome with explained variance of 49%: neurological impairment (National Institute of
Health Stroke Scale;  = 0.402, p ! 0.001), age ( = 0.233, p =
0.001), post-stroke cognitive functioning (Repeatable Battery of Neuropsychological Status, RBANS;  = 0.248, p =
0.001) and prestroke personal ADL (Barthel Index;  = 0.217,
p = 0.002). Further linear regression analyses of which RBANS
indexes and subtests best predicted long-term functional
outcome showed that Coding ( = 0.484, p ! 0.001) and
Figure Copy ( = 0.233, p = 0.002) raw scores at baseline
explained 42% of the variance in mRS scores at follow-up.
Conclusions: Early post-stroke cognitive functioning as
measured by the RBANS is a significant and independent
predictor of long-term functional post-stroke outcome.
Copyright 2011 S. Karger AG, Basel


A stroke can have serious physical, psychological, social and cognitive consequences, and may dramatically
alter a persons life. It increases the risk of dying, becoming institutionalized, demented, dependent in activities
of daily living (ADL) or depressed, and accounts for more
Jrgen Wagle
Department of Geriatric Medicine, Oslo University Hospital, Ullevaal Hospital
PO Box 4956 Nydalen
NO0424 Oslo (Norway)
Tel. +47 22 118 704, E-Mail

hospital and rest-home bed days than any other condition

[1]. However, research shows that 3042% of stroke patients have a favourable post-stroke outcome and continue to live an independent life without long-lasting, restraining symptoms [24]. It would appear important,
therefore, to identify factors that predict functional outcome in order to set realistic targets for rehabilitation and
to guide patients and their relatives accordingly. A large
number of studies have been carried out with more than
150 demographic, radiological, neurophysiological and
neurological variables reported to predict outcome [5].
Due to heterogeneity in design and setting, comparison
of results across studies is difficult, but among the most
substantiated predictors are age, initial severity of the
stroke, previous stroke, urinary continence, consciousness at onset, disorientation in time and place, severity of
paralysis, sitting balance, admission ADL score, and level of glucose outside the infarct area in hypertensive patients. In addition, female sex, diabetes, hypertension,
atrial fibrillation, physical inactivity, cardiac diseases, depression, cognitive impairment, perceived lack of social
support and absence of specialized stroke unit care have
been reported to increase the likelihood of a negative
functional outcome [58].
A limitation of many of these studies is that the findings are based on simple associations between the predictor variable and the outcome measure. However, the important question, in terms of prediction, is whether a given factor remains an independent predictor for outcome
when other known predictors are taken into account [4].
Although strong associations between early post-stroke
cognitive performance and long-term functional outcome have frequently been reported, the independent
predictive power of early post-stroke cognitive functioning is scarcely examined [1, 9]. Nys et al. [9] studied stroke
patients with a broad spectrum of neuropsychological
tests within the first 3 weeks after stroke and demonstrated that prediction of long-term functional outcome improved when specific cognitive predictors, i.e. tests of inattention and visual perceptual disorders, were added to
a standard medical model. In the review of Barker-Collo
and Feigin [1] of the impact of neuropsychological deficits on functional stroke outcomes, the authors conclude
that several cognitive domains show promise as means of
predicting post-stroke functional outcomes, including
neglect, attention deficits, verbal memory, global aphasia,
apraxia and anosognosia [1016], but emphasize that further studies are warranted. Some authors have argued
against the validity of early post-stroke cognitive testing
because of its closeness to acute effects, while other re380

Dement Geriatr Cogn Disord 2011;31:379387

searchers have found that early cognitive testing provides

valid information both with respect to long-term cognitive performance, ADL functioning, risk of institutionalization and death [1720].
The objective of this study was to evaluate functional
status at 13 months in a sample of patients referred to a
stroke rehabilitation unit and to identify prognostic factors associated with functional outcome. Specifically, we
hypothesized that cognitive functioning 24 weeks after
a stroke would predict long-term functional outcome independently of other factors.

Patients and Methods

A total of 194 stroke patients were consecutively recruited
from the Stroke Rehabilitation Unit at Ullevaal University Hospital, Oslo, Norway, during the period between March 2005 and
August 2006. Inclusion criteria were patients diagnosed with an
ischaemic or haemorrhagic stroke. Exclusion criteria were nonfluency in the Norwegian language, severe visual or hearing impairment, alcohol or drug abuse, or previously being treated for a
psychiatric illness. Thirty-one subjects (16%) withdrew from the
study during the follow up-period, leaving 163 subjects in the
study sample. Causes for withdrawal were lack of motivation to
participate (n = 25) and moving to another geographical area
(n=6). The 31 subjects who withdrew did not differ significantly
from the participating 163 patients regarding sociodemographic
or any clinical baseline variables, except for a higher proportion
of females (65%, p ! 0.05).
The Stroke Rehabilitation Unit at the Oslo University Hospital, Ullevaal, is a specialized, multidisciplinary stroke rehabilitation unit organized according to evidence-based practice for
stroke rehabilitation. It consists of experienced health professionals, including physicians, nurses, physical therapist, occupational
therapist, and speech-language therapist, who all work together
to provide an individualized rehabilitation programme tailored
to the patients specific needs. In addition, psychiatrist, social
worker and neuropsychologist services are available at the hospital and can be called for when judged needed. Due to the individualized approach in the stroke rehabilitation unit, type, intensity and duration of the rehabilitation programmes varied between the patients and were not systematically recorded or
analysed in our study. The same also goes for any stroke rehabilitation outpatient therapy received between hospital discharge and
13-month follow-up examination.
The Oslo University Hospital, Ulleval, serves a population of
approximately 180,000 people, and virtually all persons with an
acute stroke are admitted. However, only those judged to benefit
from rehabilitation are transferred to the Stroke Rehabilitation
Unit, and only when their medical conditions are stabilized. In
practice, this means that time from hospitalization to admittance
to the stroke rehabilitation unit may vary from patient to patient,
and that patients with minor or very severe strokes usually are not

Wagle /Farner /Flekky /BruunWyller /

Sandvik /Fure /Stensrd /Engedal

All subjects underwent a comprehensive structured examination while hospitalized in the stroke rehabilitation unit and were
subsequently re-examined on average 13 months later. The mean
time between hospitalization and baseline testing was 18.3 days
(SD 13.4), while follow-up examinations were undertaken at a
mean of 408.6 days (SD 41.2) later. For subjects living at home, the
follow-up examinations were carried out in the hospital, while
subjects living in nursing homes were examined at the nursing
home wards. All data collection was undertaken by 2 psychologists, L.F. and J.W.
Outcome Assessment (the Dependent Variable)
The modified Rankin Scale (mRS) [21] was administered both
at baseline and follow-up to assess the outcome. The mRS is a
7-graded global disability scale with the following scoring: 0 (no
symptoms at all); 1 (no significant disability despite symptoms,
able to carry out all usual duties and activities); 2 (slight disability,
unable to carry out all previous activities, but able to look after
own affairs without assistance); 3 (moderate disability, requiring
some help, but able to walk without assistance); 4 (moderate/severe disability, unable to walk without assistance and unable to
attend to own bodily needs without assistance); 5 (severe disability, bedridden, incontinent, and requiring constant nursing care
and attention); 6 (dead). In a review of 50 studies, the researchers
concluded that construct/convergent validity and intra-observer
reliability are strong for the mRS, but that interobserver reliability is moderate [22]. Only moderate interobserver agreement was
confirmed in a recent study even among experienced researchers
who received training and used a structured interview [23]. Still,
the mRS is the most frequently applied functional outcome measure in stroke trials. A cut-off of ^1 or ^2 is frequently applied
to classify a favourable stroke outcome from an unfavourable one.
However, dichotomization of the scale has been criticized due to
reduced statistical power, increased risk of misclassification (type
II errors), and because choice of cut-off may affect the results [23,
24]. We therefore used the full spectrum of the mRS scale in the
statistical analysis. The scoring was done by L.F. and J.W. on the
basis of a comprehensive structured clinical interview and testing
of the subjects.
Prognostic Variables (the Independent Variables):
Demographics and Biological Data
Demographics (age in years, sex, education in years) and prestroke medical variables [number of medications on admission,
hyperlipidaemia (treated), diabetes mellitus (treated), angina pectoris (treated), atrial fibrillation, myocardial infarction, previous
stroke or transient ischaemic attack and cigarette smoking] were
collected from medical records and/or by interviewing the subjects or next of kin at the baseline examination.
The strokes were classified according to type (ischaemic
strokes, ischaemic strokes with haemorrhagic transformation
and haemorrhagic strokes) and hemispheric location (left, right,
bilateral, brainstem, cerebellar, unknown) based on results from
CT scans and evaluation of clinical symptoms.
Neurological stroke-related impairment was measured with
the National Institute of Health Stroke Scale (NIHSS) [25].
The Oxfordshire Community Stroke Project classification was
applied to categorize the stroke according to topographical location and lesion severity [26].

Early Post-Stroke Cognition in Stroke

Rehabilitation Patients

Apolipoprotein E genotyping was performed on the basis of

blood samples collected from the subjects at baseline examination
according to a method described earlier [27].
Functional Data
The Norwegian version of the Informant Questionnaire on
Cognitive Decline in the Elderly was administered to evaluate
cognitive reduction over the last 10 years preceding the stroke
[28]. Due to lack of informants or subjects unwilling to involve an
informant, 131 informant interviews were obtained. We found no
significant differences between subjects with and without informants regarding any sociodemographic or baseline clinical characteristics.
Post-stroke cognitive functioning was assessed by administering the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) form A [29] at baseline and follow-up.
The RBANS is sensitive for measuring cognitive impairment after
stroke both in an acute and rehabilitation setting [19, 30, 31]. The
RBANS was administered to all subjects, even those with aphasia/
apraxia and/or hemiplegia on the dominant side. Subjects with
severe aphasia/apraxia were scored with error on tasks they were
incapable of performing, while subjects with hemiplegia that affected the dominant arm/hand used the non-dominant hand for
tests requiring writing/drawing skills (Figure Copy and Recall,
Coding). For subjects aged 65 years and above, RBANS index
scores were calculated on the basis of age- and education-corrected norms for the geriatric population [32]. Original RBANS
norms were applied for subjects under 65 years. These norms are
corrected for age, but not for education.
Personal ADL were assessed with the Barthel Index [33] and
instrumental ADL with the Frenchay Activities Index [34]. At
baseline the data was collected by interviews with nurses in the
rehabilitation unit, at the follow-up from nurses in nursing homes
or directly from the subjects or a close relative.
Symptoms of depression were examined by the Montgomery
Aasberg Depression Rating Scale [35] based on structured interviews with the patients.
Statistical Analysis
Analyses were performed using the Statistical Program for the
Social Sciences package, version 16.0. To identify prestroke and
baseline predictors of stroke outcome at 13 months, bivariate and
multiple linear regression analyses were carried out with mRS
score at the 13-month follow-up as the dependent variable. Before
performing the linear regression analyses we examined the distribution of each variable by using histograms, quantile-quantile
and box plots. Logarithmic transformation was not considered
necessary for any variable. All variables in the bivariate analysis
with a p value !0.05 were entered one by one in the linear multivariate regression analysis (following the order of the variables 
values). All variables were checked for multicollinearity. The only
variables that correlated highly (Pearsons r 1 0.7) were the NIHSS
total scale and Barthel Index at baseline.
Three predictive models were built, one including prestroke
and baseline biological data as independent variables (the biological model), one including prestroke and baseline functional data
(the functional model) and, finally, one combined model including
both biological and functional data (the combined model).
The following variables were analysed in the biological model: NIHSS total score (neurological impairment), Oxfordshire

Dement Geriatr Cogn Disord 2011;31:379387


Table 1. Basic characteristics of the subjects (n = 163)

Prestroke functioning
Informant questionnaire on
cognitive decline in the elderly
Barthel Index
Frenchay Social Activity Index

3.2480.34 3.004.92

Stroke characteristics
134 (82)
26 (16)
Ischaemic with haemorrhagic
3 (2)
Oxfordshire Community Stroke Project classification
36 (22)
Total anterior circulation syndrome
64 (39)
Partial anterior circulation syndrome
50 (31)
Lacunar syndrome
Posterior circulation syndrome
13 (8)
Lesion site
69 (42)
67 (41)
Other (bilateral, brainstem, cerebellar,
27 (17)
Neurological impairment
NIHSS total score at baseline
Baseline functioning
RBANS index scores
Immediate memory
Delayed memory
Total scale
Barthel Index
Symptoms of depression
Montgomery Aasberg Depression
Rating Scale


Ethical and Juridical Considerations

The Regional Committee for Ethics in Medical Research in
Norway and the Privacy Ombudsman approved the study protocol. Oral and written informed consent was obtained from the








Results are expressed as means 8 SD and ranges or numbers

with percentages in parentheses.

Community Stroke Project classification (topographical location/severity), lesion type (ischaemic, haemorrhagic, ischaemic
with haemorrhagic transformation), lesion site (right, left, other), prestroke vascular risk factor [previous hypertension (treated), previous hyperlipidaemia (treated), previous diabetes mellitus (treated), previous angina pectoris (treated), previous atrial
fibrillation, previous myocardial infarction, previous stroke,
previous transient ischaemic attack and cigarette smoking],
number of medications on hospital admission and genetic vulnerability through the presence of apolipoprotein E e4 allele (yes
vs. no).


In the functional model, the following variables were examined: prestroke ADL functioning (Barthel Index total score and
Frenchay Activity Index score), prestroke cognitive impairment
(Informant Questionnaire on Cognitive Decline in the Elderly
total score), baseline ADL functioning (Barthel Index total score),
baseline cognitive functioning (RBANS total index) and baseline
symptoms of depression (Montgomery Aasberg Depression Rating Scale total score).
All significant variables of the biological and functional models were used as the combined model.
In all 3 models, age, sex and years of education were included
in the analyses.
Finally, to identify specific cognitive predictors of stroke outcome at 13 months, 2 additional models were built: one with
RBANS index scores and one with RBANS subtest raw scores at
baseline as independent variables.

Dement Geriatr Cogn Disord 2011;31:379387


Subjects Characteristics
The mean age of the patients was 76.9 years (SD 10.7),
75 subjects (46%) were females, and the mean duration of
education was 11.5 years (SD 3.7). Of them 109 (67%) suffered a stroke for the first time, whereas 54 (33%) had suffered a stroke before. Characteristics of the subjects at
baseline are presented in table1.
Functional Outcome at 13 Months of Follow-Up
At the 13-month follow-up examination, 14 (9%) subjects were assigned an mRS score of 0 (no symptoms), 20
(12%) score 1 (no significant disability), 21 (13%) score 2
(slight disability), 35 (21%) score 3 (moderate disability),
28 (17%) score 4 (moderately severe disability), 8 (5%)
score 5 (severe disability) and 37 (23%) score 6 (dead).
Compared to the mRS scores at baseline, 50 subjects
(31%) had improved, 42 subjects (26%) remained stable,
and 71 subjects (44%) had worse mRS scores at followup.
Of the 126 subjects alive at follow-up, 35 lived in nursing homes and 91 at home. Except for 1 subject, all living
in nursing homes had an mRS score 12, and compared to
patients living at home they scored significantly worse on
mRS at follow-up [mean mRS score of subjects at nursing
homes 3.9 (SD 0.8) vs. mean mRS score of subjects at
home 2.0 (SD 1.3), t(124) = 8.30, p ! 0.001].

Wagle /Farner /Flekky /BruunWyller /

Sandvik /Fure /Stensrd /Engedal

Predictors of Functional Outcome at 13 Months of

The results of the unadjusted and adjusted linear regression analyses of the 3 models are presented in table2.
In the biological model, the strongest predictor of
functional outcome at 13 months was NIHSS total score
at baseline ( = 0.499; p ! 0.001), while the strongest predictor in the functional model was Barthel total score at
baseline ( = 0.393; p ! 0.001). As NIHSS total score and
Barthel Index total score correlated highly (Pearsons r =
0.804), it was decided to run these two predictors in two
separate multivariate analyses in the combined model.
When including Barthel Index (and excluding NIHSS total score), the result was identical with the result of the
functional model. The results of the combined model including NIHSS total score (and excluding Barthel Index)
are shown in table2.
Thirteen subjects (8%) experienced a new stroke during the follow-up period. In order to eliminate the effect
of a new stroke on functional outcome at 13 months, we
performed an additional multivariate linear regression
analysis (combined model) excluding the subjects with a
new stroke. The result did not change (NIHSS total scale,
baseline  = 0.421, p ! 0.001; RBANS total index, baseline
 = 0.229, p = 0.003; age in years, baseline  = 0.224, p=
0.002; Barthel Index, prestroke  = 0.224, p = 0.001; explained variance 49%).
To illustrate the numerical influence of the significant
predictors in the combined model on the mRS scores, we
calculated their scores at baseline and related these scores
to the mRS scores (table3).
Relations between Cognitive Functioning Early after
Stroke and Functional Outcome at 13 Months
As overall cognitive level at baseline turned out to be
a significant predictor of long-term stroke outcome, we
wanted to explore this relation further. Table4 presents
results of the unadjusted and adjusted linear regression
analyses with mRS score at 13 months as the dependent
variable and RBANS indexes and subtests entered as independent variables, respectively. The unadjusted analyses showed that all baseline RBANS indexes and subtests
were individually highly associated with mRS score at 13
months, but only the Visuospatial/Constructional index
(p ! 0.001), and the 2 subtests Figure Copy (p = 0.002) and
Coding (p ! 0.001) remained significant in the adjusted

Early Post-Stroke Cognition in Stroke

Rehabilitation Patients

Table 2. Predictors of mRS outcome at 13 months (n = 163)

Linear regression analysis

Biological model
NIHSS total scale
Age at baseline
OCSP classification
Number of medications on
hospital admission
Atrial fibrillation (treated)
Education years
Diabetes mellitus (treated)
Explained variance
Functional model
Barthel Index at baseline
RBANS total index at baseline
Age at baseline
MADRS at baseline
Barthel Index before stroke
Frenchay Social Activity Index
before stroke
IQCODE before stroke
Education years
Explained variance



0.493 <0.001
0.364 <0.001
0.326 <0.001

0.506 <0.001
0.351 <0.001









Combined model
0.493 <0.001
NIHSS total scale at baseline
0.364 <0.001
Age at baseline
RBANS total index at baseline 0.450 <0.001
0.334 <0.001
Barthel Index before stroke
Explained variance

0.326 <0.001
0.200 0.007
0.244 0.001
0.176 0.014



0.402 <0.001
0.233 0.001
0.248 0.001
0.217 0.002

A positive  value indicates a positive relationship between the

prognostic variable and the mRS score, and a negative  value a
negative relationship. OCSP = Oxfordshire Community Stroke
Project; MADRS = Montgomery Aasberg Depression Rating Scale;
IQCODE = Informant Questionnaire on Cognitive Decline in the


The functional long-term outcome differed markedly

among the patients. Thirteen months after stroke, only
34% of the patients could be defined as having a favourable outcome (no symptoms to slight disability), 43% of
the patients were classified as having moderate-to-severe
disability, and 23% were dead. Our results are in line with
figures from previous in-hospital and population-based
studies which have examined functional long-term outcome between 6 and 12 months after stroke using the
mRS [3638].
Dement Geriatr Cogn Disord 2011;31:379387


Table 3. Distribution of mRS scores (means 8 SD, with ranges) at 13 months of follow-up in relation to scores of significant predictor

variables at baseline (n = 163)

Baseline predictors

mRS score at 13 months of follow-up

(n = 14)

no significant
(n = 20)

(n = 21)

(n = 35)

severe disability
(n = 28)

(n = 8)


Age at baseline, years








NIHSS total at baseline








RBANS total at baseline








Barthel index before stroke








The multiple linear regression analysis revealed that a

combined model using sociodemographic, biological and
functional baseline variables best predicted functional
outcome at 13 months with explained variance of 49%
(table3). This result accords with previous results showing multiple-factor models to be the best predictors of
long-term stroke outcome [9]. Our results are also in line
with previous studies showing that severity of neurological impairment, age and prestroke ADL functioning are
major determinants affecting post-stroke outcome [7,
3942]. Neurological stroke-related impairment, as measured by NIHSS, may be viewed as a direct expression of
stroke severity, and in previous studies, a baseline score
^6 points was strongly associated with a favourable
functional prognosis at 3 months [40] and a score ^5
highly increasing the chance of being discharged to home
rather than to a rehabilitation or nursing facility [43].
With respect to age, there are several explanations why
younger patients probably have a better functional prognosis, including less comorbidity, fewer post-stroke complications, and a better physical and cognitive restorative
capacity. On these grounds, it has been suggested that age
per se should not be considered isolated in issues related
to rehabilitation and prognosis [44].
Cognitive functioning at baseline (RBANS total index) was found to be an independent predictor of longterm functional outcome. To our knowledge, this is the
first study which has used mRS scores as outcome measure in evaluating the impact of RBANS on long-term
functional outcome, and our results are therefore not

Dement Geriatr Cogn Disord 2011;31:379387

(n = 37)

easy to compare with results of previous studies. However, earlier studies of stroke rehabilitation inpatients did
find RBANS index scores to predict functional outcome
at hospital discharge [45], self-reported cognitive disability at 6 months [19], and self-reported cognitive disability,
instrumental ADL functioning and participation restrictions in domains of physical independence, mobility, occupation, social integration and economic self-sufficiency 12 months after stroke [30]. Although none of these
studies examined the independent effect of RBANS performance on long-term outcome, they still attest to the
predictive utility of RBANS in stroke rehabilitation settings.
Since cognition in the early phase after a stroke proves
to be a powerful predictor of long-term functional outcome, it is important to understand the nature of this
relationship. In the present study, all RBANS indexes
and subtests were significantly associated with outcome
in the unadjusted linear regression analyses. However, in
the adjusted analyses, only the Visuospatial/Constructional index and the 2 subtests Coding (part of the Attention index) and Figure Copy (part of the Visuospatial/Constructional index) remained significant (table4). This in general accords with Larson et al. [19], who
found performance on the Visuospatial/Constructional
index at a subacute stage to be the only RBANS measure
to correlate significantly with handicap, motor function
and frequency of activity at 6 months. Unfortunately,
these researchers did not carry out an analysis of subtests.
Wagle /Farner /Flekky /BruunWyller /
Sandvik /Fure /Stensrd /Engedal

Table 4. Cognitive predictors of mRS outcome at 13 months (n =

Linear regression analysis


RBANS indexes
Immediate memory
Delayed memory
Explained variance


RBANS subtests
List Learning
Story Memory
Figure Copy
Line Orientation
Picture Naming
Semantic Fluency
Digit Span
List Recall
List Recognition
Story Memory
Figure Recall
Explained variance











Coding and Figure Copy are dependent on motor dexterity and speed for their final execution; they are penand-pencil type of tasks. They are also visuospatial in
nature. Compromised motor and visual functioning,
ADL and symptom load as an effect of primarily righthemisphere lesions could explain the correlations. This,
however, was not brought out in a regression analysis
where motor dysfunction and right hemisphere stroke
were adjusted for (data not shown). The implication appears to be that there is a functional specificity to the tests
in question, at least beyond the motor execution part, covarying with increasing symptom load at follow-up.
Based on a functional decomposition of the 2 tests, visuospatial attention would appear to be a common denominator [46, 47]. This type of function is likely to be especially vulnerable to the kind of vascular insult induced by
the stroke, since vast territories of parietal and prefrontal
regions are affected. Especially the right-hemisphere expanses of these regions comprise central parts of the
mechanisms responsible for activation and attention [48
50]. The functional integrity of these areas would also be
affected by left-hemisphere strokes. As the severity of the
Early Post-Stroke Cognition in Stroke
Rehabilitation Patients

stroke increases, so does the risk of recruiting visuospatial dysfunction as well as mRS type of symptoms.
Our results may have implications for which cognitive
test instruments should be recommended in clinical and
research settings of stroke patients. Broad screening measures, like the Mini Mental State Examination, are often
the preferred choice, but have been criticized for being
insensitive to varying impairments, including visuospatial/constructional and executive functions [51]. We
would suggest that both Coding and Figure Copy could
beneficially be added in clinical settings, especially as
both tests are easy and quick to administer. We further
suggest that more complex neuropsychological test batteries like RBANS should be used in research to study
long-term functional outcomes.
Our study has limitations. The outcome variable mRS
is a summary tool that assesses physical disabilities with
less emphasis on cognition, emotion and social activity
[52]. Our sample is of moderate size and recruited from a
single Norwegian hospital-based stroke rehabilitation
unit, limiting its relevance for other settings. The followup period was restricted to 13 months. Future longitudinal studies should preferably follow stroke patients over
a longer time period. The number of subjects was too
small to examine differences in the level of predictive value in subgroups of the study sample, e.g. male/female,
type of stroke (ischaemic/haemorrhagic), topographical
location/severity (Oxfordshire Community Stroke Project classification), lesion site (right/left). Furthermore,
RBANS has not been validated for Norwegian-speaking
stroke patients. However, preliminary validation of
RBANS in a Norwegian psychiatric population has shown
the test battery to have adequate psychometric properties [53].
On the strong side, the study patients were recruited
consecutively and with limited use of exclusion criteria.
Based on their sociodemographic characteristics, they
are representative of older stroke patients fulfilling inclusion criteria for rehabilitation at a university hospital.
Only two administrators of tests and scales have been involved, both adhering strictly to instructions.

This paper was supported by the Eastern Norway Regional
Health Authority and the Norwegian Centre for Dementia Research, Centre for Ageing and Health.

Dement Geriatr Cogn Disord 2011;31:379387


1 Barker-Collo S, Feigin V: The impact of neuropsychological deficits on functional stroke
outcomes. Neuropsychol Rev 2006; 16: 53
2 Bonita R, Solomon N, Broad JB: Prevalence
of stroke and stroke-related disability. Estimates from the Auckland stroke studies.
Stroke 1997; 28:18981902.
3 Demchuk AM, Tanne D, Hill MD, Kasner
SE, Hanson S, Grond M, Levine SR: Predictors of good outcome after intravenous tPA
for acute ischemic stroke. Neurology 2001;
4 Counsell C, Dennis M: Systematic review of
prognostic models in patients with acute
stroke. Cerebrovasc Dis 2001; 12:159170.
5 Kwakkel G, Wagenaar RC, Kollen BJ, Lankhorst GJ: Predicting disability in stroke a
critical review of the literature. Age Ageing
6 Eriksson M, Norrving B, Terent A, Stegmayr
B: Functional outcome 3 months after stroke
predicts long-term survival. Cerebrovasc Dis
7 Demchuk AM, Buchan AM: Predictors of
stroke outcome. Neurol Clin 2000; 18: 455
8 Jongbloed L: Prediction of function after
stroke: a critical review. Stroke 1986; 17:765
9 Nys GM, van Zandvoort MJ, de Kort PL, van
der Worp HB, Jansen BP, Algra A, de Haan
EH, Kappelle LJ: The prognostic value of domain-specific cognitive abilities in acute
first-ever stroke. Neurology 2005; 64: 821
10 Paolucci S, Antonucci G, Pratesi L, Traballesi
M, Lubich S, Grasso MG: Functional outcome in stroke inpatient rehabilitation: predicting no, low and high response patients.
Cerebrovasc Dis 1998;8:228234.
11 Paolucci S, Antonucci G, Gialloreti LE, Traballesi M, Lubich S, Pratesi L, Palombi L: Predicting stroke inpatient rehabilitation outcome: the prominent role of neuropsychological disorders. Eur Neurol 1996; 36:
12 Robertson IH, Ridgeway V, Greenfield E,
Parr A: Motor recovery after stroke depends
on intact sustained attention: a 2-year follow-up study. Neuropsychology 1997; 11:
13 Jehkonen M, Laihosalo M, Kettunen JE: Impact of neglect on functional outcome after
stroke: a review of methodological issues and
recent research findings. Restor Neurol Neurosci 2006;24:209215.
14 Jehkonen M, Laihosalo M, Kettunen J:
Anosognosia after stroke: assessment, occurrence, subtypes and impact on functional
outcome reviewed. Acta Neurol Scand 2006;


15 Sundet K, Finset A, Reinvang I: Neuropsychological predictors in stroke rehabilitation. J Clin Exp Neuropsychol 1988; 10:363
16 Gialanella B, Monguzzi V, Santoro R, Rocchi
S: Functional recovery after hemiplegia in
patients with neglect: the rehabilitative role
of anosognosia. Stroke 2005; 36:26872690.
17 Patel MD, Coshall C, Rudd AG, Wolfe CD:
Cognitive impairment after stroke: clinical
determinants and its associations with longterm stroke outcomes. J Am Geriatr Soc
18 Van Zandvoort MJ, Kessels RP, Nys GM, de
Haan EH, Kappelle LJ: Early neuropsychological evaluation in patients with ischaemic
stroke provides valid information. Clin Neurol Neurosurg 2005; 107:385392.
19 Larson EB, Kirschner K, Bode RK, Heinemann AW, Clorfene J, Goodman R: Brief
cognitive assessment and prediction of functional outcome in stroke. Top Stroke Rehabil
20 Tatemichi TK, Desmond DW, Stern Y, Paik
M, Sano M, Bagiella E: Cognitive impairment after stroke: frequency, patterns, and
relationship to functional abilities. J Neurol
Neurosurg Psychiatry 1994;57:202207.
21 Van Swieten JC, Koudstaal PJ, Visser MC,
Schouten HJ, van Gijn J: Interobserver agreement for the assessment of handicap in
stroke patients. Stroke 1988; 19:604607.
22 Banks JL, Marotta CA: Outcomes validity
and reliability of the modified Rankin scale
implications for stroke clinical trials: a literature review and synthesis. Stroke 2007;38:
23 Quinn TJ, Dawson J, Walters MR, Lees KR:
Exploring the reliability of the modified
Rankin scale. Stroke 2009; 40:762766.
24 Bath PM, Gray LJ, Collier T, Pocock S, Carpenter J: Can we improve the statistical analysis of stroke trials? Statistical reanalysis of
functional outcomes in stroke trials. Stroke
25 Goldstein LB, Bertels C, Davis JN: Interrater
reliability of the NIH stroke scale. Arch Neurol 1989; 46:660662.
26 Bamford J, Sandercock P, Dennis M, Burn J,
Warlow C: Classification and natural history
of clinically identifiable subtypes of cerebral
infarction. Lancet 1991;337:15211526.
27 Wagle J, Farner L, Flekkoy K, Wyller TB,
Sandvik L, Eiklid KL, Fure B, Stensrod B,
Engedal K: Association between ApoE epsilon4 and cognitive impairment after stroke.
Dement Geriatr Cogn Disord 2009; 27: 525
28 Jorm AF: The Informant Questionnaire on
cognitive decline in the elderly (IQCODE): a
review. Int Psychogeriatr 2004;16:275293.

Dement Geriatr Cogn Disord 2011;31:379387

29 Randolph C, Tierney MC, Mohr E, Chase

TN: The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS):
preliminary clinical validity. J Clin Exp Neuropsychol 1998; 20:310319.
30 Larson E, Kirschner K, Bode R, Heinemann
A, Goodman R: Construct and predictive
validity of the repeatable battery for the assessment of neuropsychological status in the
evaluation of stroke patients. J Clin Exp Neuropsychol 2005;27:1632.
31 Wilde MC: The validity of the repeatable battery of neuropsychological status in acute
stroke. Clin Neuropsychol 2006;20:702715.
32 Duff K, Patton D, Schoenberg MR, Mold J,
Scott JG, Adams RL: Age- and educationcorrected independent normative data for
the RBANS in a community dwelling elderly
sample. Clin Neuropsychol 2003; 17: 351
33 Mahoney FI, Barthel DW: Functional evaluation: the Barthel Index. Md State Med J
34 Schuling J, de Haan R, Limburg M, Groenier
KH: The Frenchay Activities Index: assessment of functional status in stroke patients.
Stroke 1993; 24:11731177.
35 Montgomery SA, Asberg M: A new depression scale designed to be sensitive to change.
Br J Psychiatry 1979;134:382389.
36 Appelros P, Nydevik I, Viitanen M: Poor outcome after first-ever stroke: predictors for
death, dependency, and recurrent stroke
within the first year. Stroke 2003; 34: 122
37 Weisscher N, Vermeulen M, Roos YB, de
Haan RJ: What should be defined as good
outcome in stroke trials: a modified Rankin
score of 01 or 02? J Neurol 2008; 255: 867
38 Urwyler SA, Schuetz P, Fluri F, Morgenthaler
NG, Zweifel C, Bergmann A, Bingisser R,
Kappos L, Steck A, Engelter S, Muller B,
Christ-Crain M, Katan M: Prognostic value
of copeptin: one-year outcome in patients
with acute stroke. Stroke 2010;41:15641567.
39 Johnston KC, Connors AF Jr, Wagner DP,
Haley EC Jr: Predicting outcome in ischemic
stroke: external validation of predictive risk
models. Stroke 2003; 34:200202.
40 Adams HP Jr, Davis PH, Leira EC, Chang
KC, Bendixen BH, Clarke WR, Woolson RF,
Hansen MD: Baseline NIH Stroke Scale
score strongly predicts outcome after stroke:
a report of the Trial of Org 10172 in Acute
Stroke Treatment (TOAST). Neurology
41 Johnston KC, Connors AF Jr, Wagner DP,
Knaus WA, Wang X, Haley EC Jr: A predictive risk model for outcomes of ischemic
stroke. Stroke 2000; 31:448455.

Wagle /Farner /Flekky /BruunWyller /

Sandvik /Fure /Stensrd /Engedal

42 Weimar C, Konig IR, Kraywinkel K, Ziegler

A, Diener HC: Age and National Institutes of
Health Stroke Scale Score within 6 h after
onset are accurate predictors of outcome
after cerebral ischemia: development and
external validation of prognostic models.
Stroke 2004; 35:158162.
43 Schlegel D, Kolb SJ, Luciano JM, Tovar JM,
Cucchiara BL, Liebeskind DS, Kasner SE:
Utility of the NIH Stroke Scale as a predictor
of hospital disposition. Stroke 2003; 34: 134
44 Bagg S, Pombo AP, Hopman W: Effect of age
on functional outcomes after stroke rehabilitation. Stroke 2002; 33:179185.

Early Post-Stroke Cognition in Stroke

Rehabilitation Patients

45 Terryberry-Spohr L, Gordon K, List Kalnins

T: Prediction of functional outcomes in the
acute rehabilitation setting using the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS). Arch Clin Neuropsychol 2000; 15:663664.
46 Lezak MD: Neuropsychological Assessment,
ed 4. Oxford, Oxford University Press, 2004.
47 Strauss E, Sherman EMS, Spreen O: A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary, ed
3. Oxford, Oxford University Press, 2006.
48 Posner MI, Petersen SE: The attention system of the human brain. Annu Rev Neurosci
49 Liu T, Slotnick SD, Serences JT, Yantis S: Cortical mechanisms of feature-based attentional control. Cereb Cortex 2003;13:13341343.
50 Egner T, Monti JM, Trittschuh EH, Wieneke
CA, Hirsch J, Mesulam MM: Neural integration of top-down spatial and feature-based
information in visual search. J Neurosci

51 Nys GM, van Zandvoort MJ, de Kort PL, Jansen BP, Kappelle LJ, de Haan EH: Restrictions of the Mini-Mental State Examination
in acute stroke. Arch Clin Neuropsychol
52 Kasner SE: Clinical interpretation and use of
stroke scales. Lancet Neurol 2006; 5: 603
53 Lberg EM, Langeland M, Jrgensen HA:
Reliability and validity of a Norwegian
translation of the Repeatable Battery for the
Assessment of Neuropsychological Status in
patients with psychosis (poster). 10-year Anniversary of the Norwegian Neuropsychological Association, Oslo, August 2006.

Dement Geriatr Cogn Disord 2011;31:379387


Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.